基于小波变换的光寻址电位传感器信号去噪研究  

Denoising of LAPS signal based on wavelet transform

作　　者：李学亮[1] 刘诗斌[1] 陈东[1] 秋颂松[1] 

机构地区：[1]西北工业大学电子信息学院,陕西西安710072

出　　处：《光电子．激光》2014年第5期835-839,共5页Journal of Optoelectronics·Laser

基　　金：国家自然科学基金(60874101);高等学校博士学科点专项科研基金(20126102110031)资助项目

摘　　要：基于(LAPS)(光寻址电位传感器)技术的生化传感器中的光生电流是一种微弱的非平稳信号,信噪比(SNR)低。为了提取清晰的LAPS信号,且鉴于传统的傅里叶方法去噪后信号失真严重,本文采用小波变换的方法对LAPS信号进行去噪处理。通过小波变换将信号分解为3层,得到各层的小波系数以及阈值。根据每一层系数特点,按阈值进行分别处理,得到新的小波系数。最后根据新的小波系数,重构信号。对去噪后的信号进行频谱分析发现,信号频谱为有效的LAPS信号谱段。将傅里叶去噪和小波去噪方法进行对比发现,小波去噪得到信号的SNR和平滑度(SR)要高于傅里叶去噪,表明小波变换是LAPS信号去噪的有效方法。The light current of biochemical sensors based on the light addressable potentiometric sensor （LAPS） technology is a kind of weak non-stationary signal with low signal-to-noise ratio （SNR）. Hence,in order to extract clear photocurrent signal,this article adopts the method of wavelet transform for denoising processing. The LAPS signal is decomposed into three layers by wavelet transform with the wavelet coefficients and threshold of each layer. The new wavelet coefficients are obtained after process- ing the original ones according to the threshold and the characteristics of each layer. Finally, the signal is reconstructed according to the new coefficients. We find that the new LAPS signal spectrum is valid. The SNR and smoothness of the LAPS signal obtained from wavelet denoising are higher than those from the Fourier denoising,which indicates that the wavelet transform is an effective method for LAPS signal de- noising.

关 键 词：光寻址电位传感器(LAPS) 小波去噪 光电流 平滑度(SR) 

分 类 号：TP212.3[自动化与计算机技术—检测技术与自动化装置]